Measuring, Characterizing, and Reporting Pavement Roughness of Low-Speed and Urban Roads (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

80 This attachment describes proposed changes to three AASHTO specifications: • ASHTO M 328-14 Standard Equipment Specification for Inertial Profiler • AASHTO R 56-14 Certification of Inertial Profiling Systems • AASHTO R 57-14 Operating Inertial Profiling Systems These proposed changes to AASHTO M328-14, R56-14, and R57-14 are the suggestions of the NCHRP Project 10-93 staff. These specifications have not been approved by NCHRP or any AASHTO committee nor formally accepted for the AASHTO specifications. In most cases, the attachment iden- tifies changes to existing sections within AASHTO specifica- tions by underlining the added or changed portions of the text. When major changes to a section are recommended, both the current and recommended versions are shown. The following is a listing of the recommendations that motivated the proposed changes and which sections were affected: • Automatic identification and marking of adverse measure- ment conditions: M 328-14, Section 5.3.4.3 • Dynamic testing to identify specific limitations of a given profiler: R 56-14, Sections 8.1, 8.2.1, 8.2.4, 8.2.5, 8.2.5.1, 8.2.5.2, 8.2.5.3, 8.3.3, 8.3.4, 8.3.5, 8.3.6, and 8.5.1 • Testing a control section at two speeds to find mea- surement errors that often go undetected: R 57-14, Section 5.3.4.1 • Callout of the broad need for measurement of very rough roads and measurement at low speed: M 328-14, Section 4.1 • Undistorted phase response and retention of details within the profile: M328-14, Section 4.2.2 • GPS data synchronized with the DMI: M 328-14, Sec- tion 5.1 part 2 AASHTO M 328 Section 4.1, with recommended additional text underlined: 4.1 General—The equipment shall function independently from the vehicle suspension dynamics and vehicle speed throughout the operating range of 20 to 70 mph for high-speed profilers and less than 25 mph for low-speed profilers. For applications that require measurement of urban and low-speed roadways, valid operation at speeds below 20 mph and profile measurement errors over a minimal range caused by braking or stopping of the host vehicle is preferred. Section 4.2.2, with recommended additional text underlined: 4.2.2 The equipment shall have an undistorted response (profile amplitude error of less than 5 percent and loca- tion error of less than 17 percent of wavelength due to phase shift) for all wavelengths between 1.0 and 150 ft when operated between 20 and 70 mph for high-speed profilers and up to 15 and 25 mph for low-speed profil- ers. It shall also have a minimum of a 30 percent reduc- tion in profile amplitude for wavelengths shorter than 0.5 ft and longer than 300 ft. [These are the typical set- points for filters currently provided on the equipment.] The filters will have a minimum of a 70 percent reduc- tion for wavelengths shorter than 0.3 ft and longer than 450 ft. For applications that require the identification of specific roadway features or diagnosis of the sources of roughness, undistorted response for all wavelengths between 0.5 and 220 ft when operated over the valid speed range of the device is recommended. Further, no minimum reduction in amplitude at wavelengths A T T A C H M E N T 1 Proposed Changes to AASHTO Specifications

81 shorter than 0.5 ft is recommended beyond that which is necessary to avoid aliasing errors. Section 5.1, with recommended additional text underlined: 5.1. General Requirements—The profiler shall meet the following requirements: 1. The profiler shall be capable of measuring profile on pavement with an IRI range of 5 to 300 in./mi for a 0.1 mi interval. For applications that require measure- ment of urban and low-speed roadways, the profiler shall be capable of measuring profile on pavement with an IRI of up to 500 in./mi for a 0.1 mi interval and over localized roughness that causes a peak value in the continuous IRI of up to 1500 in./mi with a run- ning interval of 25 ft. 2. The profiler shall measure longitudinal distance data in feet, meters, kilometers, and miles in an incre- menting or decrementing mode from a selected start- ing point and relate the longitudinal distance to any test point. The profiler shall also measure and store GPS coordinates when possible using a system with 50 percent circular error probability of 4 inches or better in open sky conditions. The system shall relate absolute position values to longitudinal distance at 0.05-mile intervals or less using data collection syn- chronized to 10 ms or better. Optionally, the equip- ment may also report in station format (especially bump locations) for ease of contractor interface. Section 5.2, part (a) with recommended additional text underlined: 5.2. Functional Requirements—The following minimum specifications shall apply to the profiler: (a) Roadway roughness testing shall be supported at vehicle speeds up to 70 mph for high-speed profilers and 25 mph for low-speed units. Collection at speed below the valid speed range of the profiler and through stop-and-go operation shall be supported with interruption in data collection and storage and no discontinuity in measurement of longitu- dinal distance. Section 5.2.5, with recommended altered text underlined: 5.2.5 Event Marker—The profiler shall have a method for marking or recording various points of interest, such as bridges, intersections, etc., or events in the mea- surement process, such as lane changes, traverse over debris, etc. The method shall be user configurable to minimize its size and maximize its effectiveness. Section 5.2.6.2, with recommended additional text underlined: 5.2.6.2 Height Sensor—The reference height of the vehicle above the pavement shall be obtained through a noncontact height sensor module. Provisions shall be made to allow an accelerom- eter to be mounted to the height sensor case. The sensor module shall communicate with and receive power from the DAS. The height sensor shall have a resolution of at least 0.002 in. and a sufficient linear measurement range to cover the vehicle suspension motion and variation in pavement elevation. The height sensor shall oper- ate at a sufficient sampling rate to provide accurate coverage at the maximum operating speed for equip- ment utilized. This sample frequency shall allow for the implementation of an anti-aliasing filter and have sufficient short-wavelength data for dropping outliers and the implementation of a tire-bridging algorithm. Outliers or readings at the limits of the sensor range shall be replaced either by a value equal to the previous reading or a value equal to the down- ward limit of the sensor range. Moisture (except free- standing), temperature, coarse pavements, or color changes in the pavement shall not affect the func- tionality of the sensor. The samples shall be stored referenced to time or distance, or both, so that the height data may be aligned with the accelerometer and position data to provide a longitudinal profile. Section 5.3.4.3, Paragraphs 1 and 2 with recommended addi- tional text underlined: 5.3.4.3. Roadway Testing—The operational computer soft- ware shall provide all necessary functions for the operator to select and perform roadway testing for a specific location. The beginning and ending points as well as any sectioning shall be automatically detected and stored along with the primary data. The testing software shall perform testing as required by the operator. The software shall also detect abnormal conditions in the test cycle and report the condition to the operator and record the location or distance range where the abnormal condition existed as a marked event. The testing software shall save the dis- placement values from the height and distance mea- suring sensors, and the acceleration values as well as the derived elevation profile values generated by the DAS in internal memory during testing and output them to the monitor and printer for review. These results may also be transferred to removable media if selected by the operator.

82 The test software shall activate the testing using the timing and control parameters stored by the test control setup software. The software shall monitor the signals to verify that the testing is being per- formed properly and indicate detectable errors. The test software shall identify and record any of the following as a marked event: (1) any location or area in which a sensor exceeded its measurement range, (2) any area in which a sensor failed to vary over a preset travel distance, (3) any area where the profiler operated below a pre-determined mini- mum valid speed, (4) any area where the profiler experienced longitudinal acceleration or decelera- tion above a pre-determined threshold level, and (5) any location where the profiler comes to a stop. The system shall alert the operator if any of these conditions are detected. The system shall mark a predetermined range surrounding high longitudinal deceleration and stops as invalid. Predetermined settings cited above shall be defined by the manu- facturer, but user adjustable. AASHTO R 56 Section 8.1 with recommended additional text underlined: 8.1. Certification frequency shall be as specified by the Owner- Agency. The inertial profiler must successfully perform and pass certification tests to establish compliance with the minimum requirements for accuracy and repeat- ability set forth in this procedure. An inertial profiler shall be recertified after any major component repairs or replacements or any change to profiler data collec- tion software as identified in R 57. Specialized testing required for high-speed profilers to be used on urban and low-speed roadways is described in Section 8.2.4, 8.2.5, and 8.3.3 through 8.3.6. Section 8.2.1, Paragraph 4 with recommended additional text underlined: 8.2.1 Test Sections— Each test section shall be at least 528 ft in length, with proper lead-in distance and a safe stopping dis- tance available. If an index other than the IRI is evalu- ated, the test section should approach four times the length of the longest wavelength of interest of that index; however the minimum length of the test sec- tion shall be 528 ft. For the specialized dynamics test- ing described in Section 8.2.4 and 8.2.5, the minimum length of 1000 ft is required. Recommended additions to Section 8.2 of Section 8.2.4 and 8.2.5: 8.2.4 For candidate inertial profilers expected to operate below the desired certification speed, select a targeted minimum valid operating speed for certification. Per- form the testing on a 1000-ft long smooth section as defined in Section 8.2.1. Five repeat runs of the candi- date inertial profiler shall be made at a target minimum valid operating speed on the designated profile trace. 8.2.5 For profilers expected to operate during braking and stop-and-go events, perform the following procedures to establish a maximum deceleration level for valid operation and identify a range to be marked as con- taminated for excessive deceleration or at stops. Per- form the testing on a 1000-ft long smooth section as defined in Section 8.2.1. 8.2.5.1 Collect three repeat runs with the candidate inertial profiler at 45 mph. Evaluate the IRI-filtered repeat- ability cross-correlations among the runs and select a representative run. Place cones at distances of 300 ft and 480 ft from the start of the test section. Instruct the profiler driver/operator to enter the test section at 45 mph, to brake starting at the first cone such that a speed of 25 mph is achieved at the second cone, and thereafter to accelerate back to 45 mph, and then con- tinue until the end of the section. Collect two more runs using the same procedure. A record of speed as a function of distance should be provided to verify that the prescribed conditions were met in each run. The net speed change of 20 mph must be achieved within 14 ft of the prescribed 180-ft length. 8.2.5.2 Collect one run with the candidate inertial profiler at 45 mph. Place a cone at a distance of 480 ft from the start of the test section and another cone upstream of it by the following distance, D, in feet: 46.79D a g( )= Where (a/g) is the targeted acceleration level nor- malized by g. (For example, Section 8.2.4.1 recom- mends a distance of 180 ft for a target acceleration of 0.26 g.) Instruct the candidate profiler driver/opera- tor to enter the test section at 45 mph, to brake start- ing at the first cone such that a speed of 25 mph is achieved at the second cone, and thereafter to acceler- ate back to 45 mph, and then continue until the end of the section. Collect two more runs using the same procedure. A record of speed as a function of distance should be provided to verify that the prescribed con- ditions were met in each run. The net speed change of 20 mph must be achieved within a total distance that is within 8 percent of the prescribed length.

83 8.2.5.3 Collect three repeat runs with the profiler at 45 mph. Evaluate the IRI-filtered repeatability cross-correlations among the runs and select a representative run. Place cones at distances of 200 and 540 ft from the start of the section. Instruct profiler driver to enter the test section at 45 mph, and to start braking at the first cone to come to a stop within 25 ft of the second cone. Remain stopped for five seconds. Thereafter acceler- ate back to 45 mph and continue at this speed until the end of the section. Collect two more runs using the same procedure. A record of speed as a function of distance should be provided to verify that the pre- scribed conditions were met in each run. Recommended additions to Section 8.3 of Section 8.3.3 through 8.3.6: 8.3.3 Verification of Operation at a Minimum Operating Speed—Evaluate repeatability and accuracy of the pro- files measured at a proposed minimum operating speed (see Section 8.2.4) using cross correlation as described in Section 8.3.1.4. On each trace, cross correlate each of the five profiles with each of the remaining four. The repeatability agreement score for each trace is the average of the ten individual cross-correlation values. A repeatability agreement score of 0.92 is required on all traces. On each trace, cross correlate each of the five profiles to the reference profile. The accuracy agree- ment score for each trace is the average of the five indi- vidual cross-correlation values. An accuracy agreement score of 0.90 is required on all traces. 8.3.4 Verification of Areas Marked as Invalid Due to Braking— For each trace, compute and overlay short interval continuous IRI plots for the three runs with braking and the representative run from operating at constant speed. (See Section 8.2.5.1.) For each run, identify the start and the end location of the segment identified by the software in the profiler as invalid. Evaluate the con- tinuous IRI plots to see if the runs with braking satis- fied the following criteria. The area marked as invalid must include any part of the range where the continu- ous IRI plots differ by more than 30 in/mi. The range must not intrude to the area where the two continuous IRI plots are similar by more than 50 ft. 8.3.5 Verification of Valid Operation at a Given Deceleration Level—Evaluate reproducibility of the profile measured at constant speed by profiles measured with a targeted deceleration level (see Section 8.2.5.2) using cross cor- relation as described in Section 8.3.1.4. On each trace, cross correlate each of the three profiles measured with deceleration to the profile collected at constant speed. The reproducibility score for each trace is the average of three individual cross-correlation values. A score of 0.92 or greater is required to establish an expectation of valid operation at the declaration level used in the testing. 8.3.6 Verification of Areas Marked as Invalid Due to Stop- and-Go Operation—For each trace, compute and over- lay short interval continuous IRI plots for the three runs with the stop and the representative run from operating at constant speed. (See Section 8.2.5.3.) For each run, identify the start and the end location of the segment identified by the software in the profiler as invalid. Evaluate the continuous IRI plots to see if the runs with the stop satisfied the following criteria. The area marked as invalid must include any part of the range where the continuous IRI plots differ by more than 30 in/mi. The range must not intrude to the area where the two continuous IRI plots are similar by more than 50 ft. Recommended addition of Section 8.5.1 to the end of Section 8.5: 8.5.1 Testing of High-Speed Profilers with Braking and with a Stop—For specialized testing prescribed in Sections 8.2.4 and 8.2.5, the following information is provided for each trace: • Minimum valid operating speed and the associated repeatability and reproducibility scores; • Tested deceleration level and the associated repro- ducibility score; • The range deemed as invalid due to braking relative to the location where braking started; • The range deemed invalid due to stop-and-go oper- ation relative to the location of the stop. AASHTO R 57 Section 5.2.1 with the recommended addition underlined: 5.2.1. Major component repairs or replacement to an iner- tial profiler that would require the recertification of the equipment include, but are not limited to, the following: • the accelerometer and its associated hardware; • the noncontact height sensor and its associated hardware; • any printed circuit board necessary for the collec- tion of raw sensor data or the processing of the iner- tial profiles; • change of host vehicle in the case of permanently mounted equipment; • change of profile data collection software; or • major repairs to host vehicle.

84 Recommended addition of the following paragraph to Section 5.3.2.4.1: 5.3.2.4.1 . . . . . . For high-speed profilers used for network-level and project-level profiling, select one control sec- tion with a maximum IRI value of 150 in./mi and free of content concentrated at specific wavelengths (e.g., avoid jointed PCC with curl and warp or AC with roughness caused by improper roller opera- tion). Conduct a series of three measurements at the highest valid speed of the profiler and three measurements at the lowest valid speed of the profiler. If safe travel over the control section is not possible over the entire valid speed range of the profiler, modify the travel speeds accordingly. A ratio of the highest speed to the lowest speed of at least 1.5 is preferred. The average IRI measured at the high speed should not differ from the average IRI at the low speed by more than 5 percent. Cross correlation of profiles measured at unlike speeds should not produce values below 0.88 for the wave- band of interest for the IRI (see AASHTO R56) or values significantly below cross correlation of pro- files measured at the speed.